Procedural and One-Year Clinical Outcomes of Long 48 mm Xience Xpedition Everolimus-Eluting Stent in Complex Long Diffuse Coronary Artery Lesions

Background and Objectives. Long 48 mm drug-eluting stents (DES) used to treat long coronary lesions decreases the number of stents needed and avoids stent overlapping. Disadvantages include difficulty in delivery and size discrepancy between proximal and distal stent landing zones. The present study analyzed the rate of procedural, immediate angiographic, and one-year clinical outcomes of long diffuse coronary artery lesions treated with 48 mm everolimus-eluting stents (EES) and compared the clinical outcomes with multiple overlapping DES. Methods. This retrospective analysis included 213 patients with 228 lesions treated with at least one 48 mm EES at 2 hospitals in Taiwan. Results. About 40.4% of the lesions had moderate-severe calcification and 20.2% had acute angulation. The mean lesion length was 49.2 ± 18.1 mm. In 161 lesions requiring a single 48 mm EES, 67.1% had a discrepancy between proximal and distal reference diameter of ≥0.5 mm and 36% had a discrepancy of ≥1.0 mm. The procedural success rate was 98.6%. Target-vessel failure (TVF) rate at 1 year was 4.2%. Cardiac death occurred in three patients. The rates of target-vessel myocardial infarction (TV-MI), target-vessel revascularization (TVR) and definite/probable stent thrombosis were 1.4%, 3.3%, and 0.9%, respectively. After adjusting patient variables by propensity score matching, no significant difference was found for cardiac death, TVF, TV-MI, and clinically driven TVR. Conclusion. Use of 48 mm EES to treat long coronary lesions in clinically and anatomically complex patients is safe and effective. In the propensity score-matched analysis, the 48 mm EES and multiple stents have comparable clinical outcomes.

J INVASIVE CARDIOL 2022;34(2):E80-E86. Epub 2022 January 20.

Key words: coronary artery disease, drug-eluting stent, patient outcome assessment

The incidence of diabetes mellitus (DM), hypertension, dyslipidemia, and other risk factors for cardiovascular disease (CVD) is increasing worldwide,^1-4 as also the incidence of coronary artery diseases and the burden of atherosclerosis.⁵ Long diffuse coronary artery lesions account for a significant proportion of percutaneous coronary interventions (PCI) in the current era.⁶ Clinical evidence supports the efficacy and safety of contemporary drug-eluting stent (DES) implantation for treating long lesions.⁷ Before 2013, the longest DES available was 38 mm in length; therefore, 2 or more stents had to be deployed with overlap to treat coronary lesions longer than 38 mm. Although the increased incidence of in-stent restenosis and stent thrombosis caused by overlapping of first-generation DES was not found in the current second-generation DES, certain concerns remained while overlapping multiple stents.^8,9 First, geographic miss involving 2 stents may increase the risk of restenosis.^10,11 Second, deploying multiple stents may increase the procedural time, contrast amount, and radiation dose. Finally, the use of multiple stents may increase the procedure cost.¹² Everolimus-eluting stent (EES) devices in 48 mm lengths became available in Taiwan in 2015. Although the use of 1 single long 48 mm stent may avoid the disadvantages of overlapping 2 shorter stents, long stents may have their unique pitfalls. First, it could be technically challenging to deliver long stents, especially in tortuous or severely calcified coronary arteries.^8,12 Second, discrepancies between reference vessel diameters (RVDs) of the proximal and distal landing zones for long stents up to 48 mm may make appropriate stent-sizing problematic.¹³ Stent-sizing according to the proximal RVD may result in oversizing for the distal landing zone and may potentially cause distal-edge dissection. On the other hand, sizing according to the distal RVD may cause malapposition for the proximal segment before postdilation and increase the risk of longitudinal deformity during the subsequent procedure. The purpose of the present study is: (1) to analyze the rate of acute device and procedural success, and one-year clinical outcomes of the 48 mm EES for the treatment of long diffuse coronary artery lesions; and (2) to compare clinical outcomes between propensity-matched single 48 mm EES and multiple overlapping second-generation DES subgroups.

Patients receiving at least one 48 mm EES (Xience Xpedition, Abbott Vascular) at 2 medical centers in Taiwan (Chang Gung Memorial Hospital and MacKay Memorial Hospital) from April 2015 to December 2016 were enrolled retrospectively for analysis in the present study. Exclusion criteria included: (1) Patients presenting with ST-segment elevation myocardial infarction or cardiogenic shock; and (2) use of a bare-metal stent as an additional stent for the target vessel.

All patients were pretreated with 300 mg of aspirin plus 600 mg of clopidogrel (or 180 mg of ticagrelor) before PCI. An active clotting time of >300 seconds was achieved using intra-arterial heparin during the procedure. All interventions were performed according to standard techniques. All patients received dual-antiplatelet therapy (DAPT), with duration according to the clinical condition and judgment of the primary care physician.

Demographics; presentation at the index procedure; DAPT and other medications; devices at the index PCI; and clinical events at follow-up were collected from the electronic medical records of the two institutes. Procedural quantitative coronary analysis (QCA) was performed using Centricity Universal Viewer (GE Healthcare, version 6.0) by an independent cardiologist, who was unaware of the clinical information. The procedural endpoint was acute success, which included device success (successful delivery and deployment of the stent without deformation or migration, and with final Thrombolysis in Myocardial Infarction [TIMI]-3 flow and residual diameter stenosis <30%) and procedural success (device success plus absence of any major adverse cardiac events during the hospital stay). The primary clinical endpoint was target-vessel failure (TVF) at 12 months, defined as a composite of cardiac death, target-vessel myocardial infarction (TV-MI), and clinically driven target-vessel revascularization (CD-TVR). Deaths due to unknown causes were considered to be of cardiac origin. TV-MI was identified according to the universal definition.¹⁴ Other secondary endpoints were definite or probable stent thrombosis, as defined by Academic Research Consortium criteria,¹⁵ and individual components of the composite primary endpoint. The study protocol was approved by the research committees of the two institutions.

To compare the clinical outcome between patients treated by using a single long 48 mm EES and multiple (≥2 DES) overlapping second-generation DESs, we performed propensity score matching adjustment for significant differences in the baseline and lesion characteristics. Eligible patients for matching were extracted from those receiving PCI in CGMH from January 2010 to December 2016. Their data, including baseline and procedural characteristics and follow-up, were collected retrospectively. In the multiple overlapping stent subgroup, the following devices were implanted: Xience V or Xience Prime (Abbott Vascular), Promus Element or Promus Premier (Boston Scientific), Resolute or Resolute Onyx (Medtronic). Pearson chi square test was used to compare the outcome between two groups.

A total of 2610 patients underwent PCI at Chang Gung Memorial Hospital and MacKay Memorial Hospital in Taiwan during the study period. Of these, 239 patients received at least one 48 mm long EES. After considering the exclusion criteria, a total of 213 patients with 228 lesions were enrolled in the analysis (Figure 1). Baseline demographics, comorbidities, and clinical presentation are summarized in Table 1. The mean age of the patients was 64.5 years, and 80.8% of the patients were male. Risk factors for cardiovascular disease were highly prevalent; more than half of the patients presented with DM. Sixty-two (29.1%) patients had chronic kidney disease, with the estimated glomerular filtration rate being <60 mL/min/1.73 m²; among these, 16 patients (7.5%) were on maintenance hemodialysis. About 9.4% (20/213) of the patients had a left ventricular ejection fraction of <40%. Nearly half of the patients (47.9%) presented with non-ST segment elevation acute coronary syndrome (NSTE-ACS).

The mean duration of DAPT was 10.5 ± 4.1 months. Statins were prescribed for more than 80% of the study population. More than 60% of the patients received beta-blockers. Over 50% of patients received angiotensin II receptor blockers, whereas about 10.8% received angiotensin-converting enzyme inhibitors (Table 2).

The key findings of coronary angiography and QCA, along with details of the PCI procedure, are summarized in Table 3. About 74.2% (158/213) of the patients had multivessel coronary artery disease and 9.6% (22/213) had left main (LM) disease. The left anterior descending (LAD) artery was the most common target vessel (50%), followed by the right coronary artery (RCA) (36.8%) and the left circumflex artery (13.1%). Moderate-to–severe calcification was seen in 40.4% (92/228) of lesions. An angulation of >45° was observed in 20.2% (46/228) of lesions. Notably, 32.9% (75/228) of lesions were chronic total occlusions. The mean diameter stenosis was 79%, the mean lesion length 49.2 ± 18.1 mm, the mean proximal RVD 3.1 ± 0.6 mm, and the distal RVD 2.2 ± 0.7 mm. In up to 90 lesions (39.5%), the distal RVD was <2.0 mm, reflecting extremely diffuse disease.

Three-fourths (159/219) of the procedures were performed using guiding catheters with stronger backup support, including extra back up or left amplatz catheter for left coronary artery and left amplatz with or without a short tip for right coronary artery. A guide extension catheter was used in 13 cases (5.5%) to facilitate delivery of the stent. Approximately one-fifth of the PCI procedures (n = 51) were performed under intravascular imaging guidance (intravascular ultrasound [IVUS] in 48 procedures and optical coherence tomography [OCT] in 3 procedures). Predilation and postdilation were performed in up to 99.6% and 96.9% of lesions, respectively. Rotablation or cutting balloon was used for lesion preparation in 4.8% of procedures. Nineteen procedures (8.3%) involved the use of 2 stents for true bifurcation lesions. The mean stent length per target lesion was 60.1 ± 20.6 mm (48-142 mm). Therefore, up to 29.3% (67/228) of lesions required 1 or 2 stents in addition to an EES of 48 mm. The mean stent diameter was 3.1 ± 1.8 mm. For lesions requiring a single 48 mm EES, size discrepancies between the proximal and distal landing zones were noted to be common. Among 161 such lesions, up to 108 (67.1%) demonstrated a discrepancy of ≥0.5 mm between the proximal and distal RVDs, and out of these 108 lesions, 58 (36%) demonstrated a discrepancy of ≥1.0 mm. For these lesions, the stent-sizing strategy was at the operator’s discretion. About 63.3% of the lesions were treated with a proximal-sizing strategy using a stent 0.5 mm larger than the distal RVD; of these lesions, 21.7% were treated using a stent 1 mm larger than the distal RVD. For lesions treated with the proximal-sizing strategy, the stent was slowly expanded with a deployment pressure lower than the nominal pressure (around 5–7 atm), which resulted in a partial expansion of the stent to a size close to the distal RVD estimated visually. The remaining 36.7% were treated with a distal-sizing strategy.

The procedural and clinical endpoints are summarized in Table 4. The device success rate was 100%. All of the stents were successfully delivered and deployed without deformation or migration. The procedural success rate was 98.6% (210/213). Two patients had residual stenosis >30% because of severe calcification, and 1 patient had final TIMI-2 flow.

At 1 year, the TVF rate was 4.2% (9/213). Cardiac death occurred in 3 patients. One patient, who was treated with 2 long stents from the LM ostium to distal LAD for extremely diffuse disease with severe calcification, experienced definite subacute stent thrombosis 6 days after PCI. Another patient, who experienced probable late-stent thrombosis, presented with sudden cardiac death (documented ventricular fibrillation) 6 months after receiving LM bifurcation stenting. One patient died of advanced heart failure. Three patients had TV-MI (1.4%). Two patients experienced definite/probable stent thrombosis (0.9%, 2/213). The rate of CD-TVR was 3.3% (7/213).

A total of 298 patients fulfilled the criteria for the propensity analysis. There were 149 patients each in the single 48 mm EES and multiple overlapping stents subgroup. The baseline, demographic and clinical characteristics of the study groups are presented in Table 5. There were no significant differences between the 2 treatment groups.

The lesion and procedural characteristics did not differ between single 48 mm EES and multiple stents groups in the propensity analysis. The mean stent length was 48.00 ± 0.00 mm and 47.79 ± 4.38 mm while the mean stent diameter was 2.93 ± 0.43 mm and 2.92 ± 0.30 mm, in the 48 mm EES and multiple stents subgroups, respectively (Table 5).

At 1-year follow-up, the incidence of TVF (P=.76), cardiac death (P=.55), TV-MI (P=.31), and CD-TVR (P=.98) were not statistically different between single 48 mm EES and multiple stents subgroups in the propensity analysis (Table 6).

In the present study, we showed that using 48 mm EES to treat complex long diffuse coronary lesions in patients with high clinical risk is associated with high acute procedural success and a low TVF rate at one year. Further, the propensity score-matched analysis showed no significant difference for TVF, cardiac death, TVMI, and CD-TVR between single 48 mm EES and multiple stents subgroups.

Current second-generation DES have demonstrated improved performance, in terms of both acute procedural and long-term clinical outcomes. The length of the stent has not been found to be associated with adverse events or to have an impact on clinical outcomes.^16,17 Hence, second-generation DES of longer lengths have been made available recently. Using longer stents may reduce the number of stents required to treat long lesions and help in avoiding overlapping of multiple stents. However, DES as long as 48 mm have the inherent pitfall of increased difficulty in delivery, especially in calcified or tortuous vessels.^8,12

Unfortunately, calcification and tortuosity are common characteristics of long diffuse coronary disease because of common risk factors, such as DM and old age.^18,19 In the present study, moderate or severe calcification was found in as high as 40.4% of all lesions and acute angulation in 20.2%. For successful stent delivery, a guiding catheter with stronger backup support was used in up to three-fourths of procedures. Further enhancement of support was provided by using a guide extension catheter in 5.5% of the procedures. In addition, rotablation or a cutting balloon was used for lesion preparation in 4.8%. With all these preparations, 100% success in delivery of the 48 mm EES was achieved in this study. A high device success rate was reported by Tan et al in a single-center, observational study using the same DES.¹³

A potential concern regarding the use of a long stent of 48 mm to treat long diffuse lesions is the discrepancy between the proximal and distal RVDs, which makes appropriate stent-sizing problematic.¹³ In the present study, among 161 lesions treated using a single 48 mm EES, up to 67.1% of lesions showed a discrepancy of ≥0.5 mm; and 36% showed a discrepancy of ≥1.0 mm between the proximal and distal RVDs. A distal-sizing strategy for lesions with a size discrepancy may result in malapposition at the proximal segment after initial stent deployment, which increases the risk of longitudinal deformation during the following procedure. In addition, overexpansion of an undersized stent during postdilation could lead to stent fracture and damage of the polymer coating, which may expose patients to the potential risks of thrombosis and restenosis.^20,21 On the contrary, a proximal stent-sizing strategy may increase the risk of distal-edge dissection. In the present study, the stent-sizing strategy for lesions with size discrepancy was decided at the operator’s discretion. Up to 63.3% of lesions were treated with a proximal-sizing strategy using a stent that was ≥0.5 mm than the distal RVD; of these lesions, 21.7% were treated using a stent 1 mm larger than the distal RVD. The distal-sizing strategy was applied for the remaining 36.7% of lesions. With meticulous deployment, no distal-edge dissection or longitudinal deformation occurred in the present study in spite of the high prevalence of size discrepancy.

Given the highly complex lesion characteristics (mean lesion length of 49 mm, distal RVD <2 mm in more than one-third of lesions) and highly complex clinical characteristics (more than one-half of patients with diabetes, nearly one-third with CKD—including 7.5% on maintenance hemodialysis and nearly half with NSTE-ACS), the one-year clinical outcomes of this study are favorable and comparable to a previous pivotal trial, ie, the SPIRIT III trial, which used a similar DES platform in patients with much shorter lesions and fewer clinical comorbidities.²² In the propensity score-matched analysis, the incidence of TVF, cardiac death, TVMI and CD-TVR did not differ between single 48 mm EES and multiple overlapping stents subgroups. Similarly, in a historical cohort study, the incidence of cardiac death (1.5% and 1.1%; P=.52), TVR (2.8% and 3.2%; P=.62), TLR (1.6% and 1.9%; P=.67) were not statistically significant between long stent (PRX [Promus/Resolute/Xience] = 38 mm) and overlap stent groups.²³ A Singaporean observational registry from tertiary care cardiac centers showed no statistical difference for cardiac death, TLR, TVR (all 0.9% and 2.0%, respectively) and TLF (2.6% vs 4.1%) between 48 mm EES and overlapping DES at 1-year follow-up.²⁴ A prospective registry¹¹ reported higher risk of composite of death, MI, and TVR in sirolimus-eluting stent (SES) with overlapping stent group than single DES group (28.7% vs 17.5%; P<.001). However, the outcome was not statistically different between DES with overlapping stent group vs single EES group (22.3% vs 20.4%; P=.20), suggesting improved performance with second-generation DES. Overall, performing single stenting with 48 mm stents has comparable clinical outcomes as multiple stents; the former intervention is cost-effective, and may decrease the procedure time, contrast and radiation dose. With the availability of 48 mm EES in Taiwan, our findings extend current knowledge on treating long coronary lesions.

Our study has several limitations. First, it is not a randomized controlled trial. Therefore, it has inherent potential bias. Second, periprocedural cardiac biomarkers and electrocardiogram were not routinely checked. Thus, the incidence of periprocedural MI could not be analyzed, which may have led to an overestimation of the procedure success rate. Third, intravascular imaging was used in a small percentage of the procedures described. Therefore, the atheroma burden of these long lesions could not be accurately evaluated. Finally, this study followed patients for only a year. Long-term follow-up is mandatory to understand the long-term performance of the long stent.

Treating long coronary lesions in clinically and anatomically complex patients with 48 mm EES is safe and effective. Further, a propensity score matched analysis indicated comparable clinical outcomes between single 48 mm EES and multiple stents. Larger, prospective, long-term studies may be required to further reinforce the findings of this retrospective analysis.

Acknowledgments. We would like to thank BioQuest Solutions Pvt Ltd for their support in data analysis and for their editorial assistance.

• A Long-Term Single-Center Registry of 6893 Patients Undergoing Elective Percutaneous Coronary Intervention With the Xience Everolimus-Eluting Stent
• Four-Year Outcomes of Multivessel Percutaneous Coronary Intervention With Xience V Everolimus-Eluting Stents
• One-Year Outcomes of Orbital Atherectomy of Long, Diffusely Calcified Coronary Artery Lesions
• Outcomes of Overlapping Bare-Metal Stents with Sirolimus-Eluting Stents for Long Lesions in Small Coronary Vessels